The proposed project is a multisite study of testing and counseling (T&C) for HIV, designed to investigate the factors that hinder or facilitate uptake in different settings. The uptake of T&C is key to the success of HIV treatment and prevention programs, and it is especially important to understand its determinants in view of current debates regarding the shift from stand-alone voluntary T&C to integrated routine services. The study makes systematic comparisons across four sites that differ in terms of the provision of T&C and the social context in which services are provided. The first phase of the analysis is carried out for each site separately, and investigates three sets of factors thought to influence the uptake of testing: the behaviors and motivations of users and non-users of services towards testing and disclosure; the provision of testing and counseling, including whether testing is client- or provider-initiated; and the implementation of practices related to confidentiality and consent. The second phase of the analysis consists of systematic comparisons across sites to identify the particular characteristics of programs and context that are associated with the uptake of testing and with positive responses among clients. Among the specific questions examined in comparative analyses are: the extent of the association of stigma with testing and disclosure; the influence of gender on testing and disclosure; the circumstances that shape practices related to informed consent and confidentiality; and the extent to which these practices influence clients' responses. The study is based on interviews with health providers, with clients of testing services, and with non- users (people who have not been tested), as well as on observations at study sites. Data will be collected at health facilities, including: stand-alone HIV testing and counseling facilities, facilities that provide care to people living with HIV, and facilities that provide HIV testing and counseling as part of providing other health services, such as antenatal care, care for tuberculosis, or for sexually transmitted infections. Analyses will combine statistical methods for quantifiable survey data, and qualitative methods for the textual data resulting from open-ended responses and observations. [unreadable] [unreadable] [unreadable]